ABSTRACT1 The overarching goal of this project is to test predictions from theories of Developmental Origins of Adult Health and Disease (DOHaD), of which the best known variant is ?Barker's hypothesis?, using very long period and cohort mortality time trends. We will use three main sources of data: the Latin American Mortality database (LAMBdA; see webLAMBdA) and the Human mortality database (HMD). In addition, we will use three longitudinal datasets of elderly people fielded in Costa Rica (CRELES), Mexico(MHAS) and Puerto Rico(PREHCO), all of which contain information on adult health and mortality as well as on early conditions. We will estimate a broad spectrum of mortality statistics to characterize child, adult, and old age mortality trajectories during the secular mortality decline, around 1780-2010 in Western-Southern and Northern Europe and North America (NAWE), and 1900-2010 in Latin American and Caribbean (LAC). The project is organized around two distinct tasks: (a) generalizing and fine-tuning an existing formal model to represent adult mortality patterns influenced by delayed effects predicted by DOHaD; (b) testing predictions using two databases, LAMBdA and HMD and CRELES, MHAS and PREHCO as ancillary data sets. 1 Abbreviations used throughout the text: LAMBdA Latin American Mortality Database<website> HMD Human Mortality Database<website> LAC Latin American and Caribbean countries NAWE North America and Western-North-South Europe webLAMBdA LAMBdA official web sitewww.ssc.wisc.edu/cdha/latinmortality DOAhD Developmental Origins of Adult Health and Disease